Tootfinder

Mitigating Audiovisual Mismatch in Visual-Guide Audio Captioning
Le Xu, Chenxing Li, Yong Ren, Yujie Chen, Yu Gu, Ruibo Fu, Shan Yang, Dong Yu
https://arxiv.org/abs/2505.22045

Mitigating Audiovisual Mismatch in Visual-Guide Audio Captioning
Current vision-guided audio captioning systems frequently fail to address audiovisual misalignment in real-world scenarios, such as dubbed content or off-screen sounds. To bridge this critical gap, we present an entropy-aware gated fusion framework that dynamically modulates visual information flow through cross-modal uncertainty quantification. Our novel approach employs attention entropy analysis in cross-attention layers to automatically identify and suppress misleading visual cues during mo…

Pixel-wise Modulated Dice Loss for Medical Image Segmentation
Seyed Mohsen Hosseini
https://arxiv.org/abs/2506.15744 https://arxiv.or…

Pixel-wise Modulated Dice Loss for Medical Image Segmentation
Class imbalance and the difficulty imbalance are the two types of data imbalance that affect the performance of neural networks in medical segmentation tasks. In class imbalance the loss is dominated by the majority classes and in difficulty imbalance the loss is dominated by easy to classify pixels. This leads to an ineffective training. Dice loss, which is based on a geometrical metric, is very effective in addressing the class imbalance compared to the cross entropy (CE) loss, which is adopt…

Low-resource keyword spotting using contrastively trained transformer acoustic word embeddings
Julian Herreilers, Christiaan Jacobs, Thomas Niesler
https://arxiv.org/abs/2506.17690

Low-resource keyword spotting using contrastively trained transformer acoustic word embeddings
We introduce a new approach, the ContrastiveTransformer, that produces acoustic word embeddings (AWEs) for the purpose of very low-resource keyword spotting. The ContrastiveTransformer, an encoder-only model, directly optimises the embedding space using normalised temperature-scaled cross entropy (NT-Xent) loss. We use this model to perform keyword spotting for radio broadcasts in Luganda and Bambara, the latter a severely under-resourced language. We compare our model to various existing AWE a…

Single-cell metabolic flux analysis reveals coexisting optimal sub-groups, cross-feeding, and mixotrophy in a cyanobacterial population
Ari\'an Ferrero-Fern\'andez, Paula Prondzinsky, Lucia Gastoldi, David A. Fike, Harrison B. Smith, Daniele De Martino, Andrea De Martino, Shawn Erin McGlynn
https://arxiv.org/abs/2506.059…

Single-cell metabolic flux analysis reveals coexisting optimal sub-groups, cross-feeding, and mixotrophy in a cyanobacterial population
We derive a single-cell level understanding of metabolism in an isogenic cyanobacterial population by integrating secondary ion mass spectrometry (SIMS) derived multi-isotope uptake measurements of Synechocystis sp. PCC6803 with a statistical inference protocol based on Liebig's law of the minimum, the maximum entropy principle, and constraint-based modeling. We find the population is structured in two metabolically distinct clusters: cells optimizing carbon yield while excessively turning over…

Robust Unsupervised Adaptation of a Speech Recogniser Using Entropy Minimisation and Speaker Codes
Rogier C. van Dalen, Shucong Zhang, Titouan Parcollet, Sourav Bhattacharya
https://arxiv.org/abs/2506.10653

Robust Unsupervised Adaptation of a Speech Recogniser Using Entropy Minimisation and Speaker Codes
Speech recognisers usually perform optimally only in a specific environment and need to be adapted to work well in another. For adaptation to a new speaker, there is often too little data for fine-tuning to be robust, and that data is usually unlabelled. This paper proposes a combination of approaches to make adaptation to a single minute of data robust. First, instead of estimating the adaptation parameters with cross-entropy on a single error-prone hypothesis or "pseudo-label", this paper pro…